TReVS: Integrating Textual Relevance and Visual Saliency for Efficient Vision-Language Model Token Pruning

arXiv:2609.37581 · cs.CV, cs.AI · Submitted 2026-09-29 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-09-29

Updated: 2026-09-29

Terminology

• Pruning — 58× in this paper · explained in 4 episodes
• Vision-language models — 8× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Redundancy — 7× in this paper · explained in 2 episodes
• Benchmark — 6× in this paper · explained in 2 episodes
• Large Language Model — 5× in this paper
• Cosine Similarity — 3× in this paper · explained in 3 episodes
• Hallucination — 3× in this paper · explained in 4 episodes
• Vision-Language Models (VLMs) — 3× in this paper · explained in 6 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Ablation study — 2× in this paper · explained in 1 episode
• Aggregation — 2× in this paper · explained in 1 episode
• Consistency — 2× in this paper · explained in 3 episodes
• Median Absolute Deviation — 2× in this paper
• Multimodal Reasoning — 2× in this paper · explained in 5 episodes
• Robustness — 2× in this paper · explained in 5 episodes
• Training-Free Framework — 2× in this paper · explained in 2 episodes
• Alignment — 1× in this paper · explained in 1 episode
• Clustering — 1× in this paper · explained in 2 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Farthest Point Sampling — 1× in this paper
• Grounding — 1× in this paper · explained in 3 episodes
• Guided attention — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• KV cache — 1× in this paper · explained in 3 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Temporal Reasoning — 1× in this paper · explained in 2 episodes
• Two-Stage Framework — 1× in this paper · explained in 2 episodes

Sources

• GPT-4 Technical Report
• Qwen Technical Report
• Qwen2.5-VL Technical Report
• Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding
• MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models
• LLaVA-OneVision: Easy Visual Task Transfer
• Evaluating Object Hallucination in Large Vision-Language Models
• Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
• LLaMA: Open and Efficient Foundation Language Models
• PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction
• MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models